Michael Uckelmann

LPT uses bounded agentic loops with LLMs and a curated database to design protein binders from natural language prompts, achieving high success rates with low computational costs.

The paper constructs an atlas of human multi-protein complexes by integrating computational structure predictions with genetic dependency data, identifying 354 high-confidence complexes including novel ones.

This study shows that H3.3K9 and K27 modifications regulate repressive chromatin states at cis-regulatory elements and bivalent promoters in mESCs, affecting gene expression.